selecting legal advisors in m&as: organizational … · selecting legal advisors in m&as 2...
TRANSCRIPT
Selecting Legal Advisors in M&As 1
SELECTING LEGAL ADVISORS IN M&AS: ORGANIZATIONAL LEARNING AND
THE ROLE OF MULTIPLICITY OF MENTAL MODELS*
Bastian Westbrock
Utrecht University
Katrin Muehlfeld
Trier University
Utrecht University
Utz Weitzel
Utrecht University
Radboud University
FORTHCOMING IN: JOURNAL OF MANAGEMENT
*Supplemental material available online: https://utzweitzel.files.wordpress.com/2017/12/online-
appendix-corr-table-a4.pdf
Acknowledgements: The authors would like to thank associate editor William P. Wan and two
anonymous reviewers for their excellent comments and guidance in improving the article. They
also thank seminar audiences at the AOM 2013 and AIB 2013 conferences for useful comments
on earlier drafts. The authors particularly thank Joanna Barth and Anja Loderer for valuable
research assistance in terms of data collection. Any remaining errors are our own.
Corresponding author: Katrin Muehlfeld, Trier University, Universitätsring 15, 54286 Trier,
Germany. Email: [email protected]
Selecting Legal Advisors in M&As 2
ABSTRACT
Management scholars have identified a variety of firm characteristics as antecedents to
organizational learning. In this study, we conceptualize intra-organizational multiplicity of
mental models as a complementary element that facilitates shifting from lower- to higher-level
learning. Specifically, we investigate whether multiplicity of mental models—proxied by four
different measures—, helps acquirers to categorically adapt selection rules for legal advisors in
M&As from domestic towards international settings. In developing our conceptual framework,
we integrate resource-based, social network, and organizational learning perspectives.
Empirically, we draw on 11,511 acquisition attempts announced in 1998–2010 (completion/
abandonment assessed as of January 10th, 2015, at the latest). The results largely support our
theory: First, choosing legal advisors in domestic and international deals calls for different
selection rules. While in domestic deals, network-related characteristics are more important
drivers of lawyers’ performance relative to their country-specific expertise, the comparative
relevance of these attributes is reversed in cross-border deals. Yet, initially, acquirers fail to
recognize this. Also, they do not initially adjust their selection criteria appropriately in response
to accumulating M&A experience of their own. Only after having attempted a substantial
number of cross-border M&As, they reach a turning point at which they rebalance their selection
criteria such that they reflect the predominant relevance of country-expertise in cross-border
settings. Finally, recognition of the need to categorically reassess selection criteria in
international deals is significantly facilitated by an acquirer’s multiplicity of mental models.
Keywords: Organizational learning; experiential learning; social networks; legal advisors; cross-
border vs. domestic M&As; acquisition completion
Selecting Legal Advisors in M&As 3
SELECTING LEGAL ADVISORS IN M&AS: ORGANIZATIONAL LEARNING AND
THE ROLE OF MULTIPLICITY OF MENTAL MODELS
INTRODUCTION
Some organizations appear to be systematically better at learning than others (Argote,
2013). What sets them apart? Argote and Miron-Spektor (2011) conceptualized the
organizational context as composed of various elements whose interplay affects an
organization’s ability to learn. Prior research has empirically identified several contextual
antecedents of organizational learning, such as organizational scope (specialist vs. generalist),
culture, absorptive capacity, aspiration level, intra- and inter-organizational network relations, IT
and knowledge management systems, and, for intra-organizational group learning processes,
lower-level constructs such as team member diversity (Argote, 2013).
Here, we propose intra-organizational multiplicity of mental models—defined as a count
of distinct mental models relating to a specific information domain within the organization—as a
complementary contextual element that may foster specific types of organizational learning. Our
arguments differ from prior research in that we propose that the impact of multiplicity of mental
models is less domain-specific than the influence of related constructs such as absorptive
capacity (Cohen & Levinthal, 1990). We suggest that it allows for ‘meta-contextual’ learning,
that is learning in and across various and even otherwise unrelated domains. In the context of
organizational learning, mental models have been characterized as “deeply held internal images
of how the world works” (Senge, 1990:163). Greater multiplicity of mental models within the
organization, first, makes available a larger range of cognitive structures for describing and
dealing with a specific type of task (e.g., sourcing advisory services). Second, we propose that
multiplicity of mental models facilitates the emergence of a generalized awareness of the
Selecting Legal Advisors in M&As 4
existence of a whole range of cognitive structures—independent of the specific task. By
highlighting, generally, the possibility of ‘doing things differently’, we maintain that multiplicity
may facilitate a specific type of learning: recognition, in response to accumulating experience, of
the need to shift from lower-level learning, which entails adaptation within a given set of rules,
to higher-level, rule-changing learning (cf. Fiol & Lyles, 1985).
In this paper, we thus seek to address whether an organization’s multiplicity of mental
models positively affects recognition of the need to shift to higher-level learning. We address
this issue in the context of a specific research setting: Acquirers’ selection of legal advisors in
mergers and acquisitions (M&As) constitutes the object of learning in this study. Given that the
relevance of different types of advisors varies across the M&A process, we zoom in on a specific
stage during which legal advisors are of particular importance, the public takeover phase
(Karsten, Malmendier, & Sautner, 2014; Krishnan & Masulis, 2013). We study if acquirers apply
selection rules, which they have developed in domestic M&As, to international deals: On the one
hand, institutional differences across countries (e.g., Dikova, Rao Sahib, & van Witteloostuijn,
2010) create strong pressures to re-assess established routines, in particular related to the role of
country-specific legal expertise. On the other hand, the credence good nature of legal services
(Coates, 2001) and the potential ‘dark side’ of embeddedness (e.g., Lee, 2013) of acquirers’
network relations with lawyers counteract these pressures.
Following research on partner selection (e.g., Gulati & Gargiulo, 1999; Hitt, Dacin,
Levitas, Arregle, & Borza, 2000; Lee, 2013), we blend resource-based, social network, and
organizational learning perspectives in deriving our hypotheses. First, we analyze acquirers’
reliance in domestic deals on relational- vis-à-vis expertise-related law firm attributes for advisor
selection and consider their respective contribution to advisor performance. Second, we
Selecting Legal Advisors in M&As 5
investigate whether the same criteria govern advisors’ performance in international M&As and
whether acquirers employ them as a basis for selection. Our basic expectation is that bidders
need to nominate advisors based on a reversed weighting of relational and expertise-related
attributes when they operate internationally, but that they do not recognize this initially. Next, we
study the effect of experiential learning: When accumulating international M&A experience, do
acquirers learn to fundamentally rebalance the selection criteria to better align them with the
drivers of law firm performance in international deals? Following Haleblian and Finkelstein’s
(1999) theory of inappropriate generalization, we expect that they will initially fail to do so, but
that sufficient experience will facilitate the shift from lower- to higher-level learning needed for
a categorical rebalancing of selection criteria. Lastly, we ask whether firms with greater
multiplicity of mental models require less experience in order to reach this turning point.
An analysis of 5,594 domestic and 5,917 cross-border M&A attempts announced in 1998–
2010 supports our theory. First, selecting legal advisors calls for distinct weighting of network-
relative to expertise-related attributes in domestic and international deals. Second, acquirers learn
from experience: There exists a turning point beyond which the typical bidder relies more on
expertise- rather than network-related selection criteria in cross-border deals. Third, multiplicity
of mental models—assessed at various organizational levels and in several domains—appears to
facilitate the required shift from lower- to higher-level learning. In sum, this study provides
important insights regarding organizational learning, social networks, and M&As, and for
managerial practice.
RESEARCH SETTING: THE ROLE OF LEGAL ADVISORS IN COMPLETING M&AS
M&A processes consist of two broad stages: pre-completion, which ends with the official
resolution, and integration (Haspeslagh & Jemison, 1991). Two critical events mark the pre-
Selecting Legal Advisors in M&As 6
completion phase (Boone & Mulherin, 2007): The private takeover period starts with the initial
screening of potential business combinations and ends with the official announcement of a
preliminary merger agreement. The ensuing public takeover phase starts with this announcement
and ends with either completion or abandonment. While a legal advisor might be involved in
drafting the preliminary agreement, the primary duty of the counsel begins after this agreement
is signed and the deal is announced (DePamphilis, 2008). The counsel is then responsible for (i)
obtaining shareholder, regulatory, and third-party consent, and (ii) structuring, drafting, and
completing the definitive agreement of purchase and sale in a way that accommodates the
diverse interests and complies with all relevant regulations of every country where acquirer and
target are listed or incorporated (Gilson & Black, 1996). Lawyers advise clients on
responsibilities in making or responding to offers, act as principal negotiators for legal terms,
provide legal due diligence, draft contracts, ensure that corporation, anti-trust, securities and tax
laws are being followed, and are responsible for M&A related litigation (Krishnan & Masulis,
2013). How can their performance during the public takeover phase be assessed?
While the release of new information may always prompt the necessity to reassess a deal
(Chakrabarti & Mitchell, 2016; Zhou, Xie, & Wang, 2016), prior literature has argued that
during this public takeover stage, completion and, in particular, closing time (i.e., the time it
takes to proceed from the initial public announcement to the definitive agreement) constitute
important outcome variables from the viewpoint of acquirers (Chakrabarti & Mitchell, 2016;
Dikova et al., 2010; Karsten et al., 2014).1 Particularly at this stage, the involved parties need to
resolve issues of legal nature in order to swiftly complete an announced M&A. M&As are
complex legal transactions because of the large number of stakeholder groups that is involved,
including managers, shareholders, banks, employees, and regulatory authorities (Boone &
Selecting Legal Advisors in M&As 7
Mulherin, 2007; DePamphilis, 2008). Against this background, the legal counsel has been
described as a “transaction cost engineer” (Gilson, 1984: 244) who has a major impact on the
likelihood that an agreement is reached and, in particular, on the time it takes to close a deal
(Gilson & Black, 1996; Krishnan & Masulis, 2013; Karsten et al., 2014). Hence, a lawyer’s
contribution to reducing closing time can be viewed as a key indicator of its performance.
THEORY AND HYPOTHESES
Acquirers’ Selection of Legal Advisors and Law Firm Performance in Domestic M&As
How do acquirers select lawyers for the public takeover phase and what drives advisors’
performance? Little direct evidence exists, but research on partner selection (e.g., Hitt et al.,
2000) including alliances (e.g., Chung, Singh, & Lee, 2000; Gulati & Gargiulo, 1999), suggests
at least two categories of selection criteria, which may also affect performance: first, social
network relations; second, access to complementary resources and capabilities.
Social network relations. The literature on partner selection in inter-organizational
networks (e.g., Gulati, 1995; Mitsuhashi & Min, 2016) and on social networks in professional
services (e.g., Baum, Rowley, Shipilov, & Chuang, 2005; Jensen, 2003) has identified network-
related attributes—also referred to as embeddedness—, such as past direct ties between
organizations as strong predictors of selection in future exchange episodes. Social networks
provide unique mechanisms for the governance of transactions and dissemination of information
(Powell & Smith-Doerr, 1994). Ties from repeat interactions between the same actors enable
mutual learning, motivate the sharing of private information, and reduce the potential for
frictions in subsequent episodes (Uzzi & Lancaster, 2004).
Studies on network relations of professional service firms (e.g., Baker, 1990; Uzzi &
Lancaster, 2004) suggest that past direct ties between clients and advisors not only matter for
Selecting Legal Advisors in M&As 8
selection but may also affect outcome variables—as would be expected from a rational
capability-based logic of partner selection. How may network relations between acquirer and
lawyer enable the legal counsel to perform better during the public takeover phase? Past
interactions can help in various ways: First, they enable direct learning about the partner’s
corporate culture and internal procedures. Familiarity with the bidder’s ways of working during
the public takeover stage, with its intra-organizational routines, organizational structure, and
decision-making bodies—both formal and informal—facilitate coordination between lawyer and
client regarding the design of the acquisition contract (Karsten et al., 2014) and, generally, the
structuring of the diverse interests and compliance with all relevant regulations. Second, past ties
facilitate the emergence of trust (Gulati, 1995; Uzzi & Lancaster, 2004); especially if
accompanied by a degree of closeness and exclusivity in the relationship (Coleman, 1990;
Gargiulo & Benassi, 2000) that goes beyond mere familiarity. Trust stimulates relationship-
specific investments, for example, in terms of inter-organizational routines in jointly handling
legal issues during the public takeover stage. Also, it may stimulate sharing of private
information (Gulati, 1995; Uzzi & Lancaster, 2004) about motives of and intentions for a
merger. Such information, in turn, enables the lawyer to better anticipate the acquirer’s concerns
during negotiations, speeding up internal discussions between bidder and advisor and, thereby,
also the process of exchanging mark-ups between the parties (Karsten et al., 2014).
Country-specific expertise. Firms tend to team up with other firms that possess
complementary resources or capabilities (e.g., Chung et al., 2000; Mitsuhashi & Min, 2016; Shah
& Swaminathan, 2008). While much research considers horizontal alliances, the argument likely
extends to acquirers’ relationships with legal advisors. Given the complexity of M&As, which
touch upon many different areas of law (Gilson & Black, 1996), and their relative rarity as
Selecting Legal Advisors in M&As 9
corporate events, most companies find it inefficient to rely strongly on in-house counsel. Instead,
they seek the expertise of law firms specializing in M&As (e.g., Heinz, Nelson, & Laumann,
2001). A broad range of legal domains is relevant in mergers and countries differ significantly in
these respects. Also, given the dispersed and highly regulated nature of the international legal
service industry (e.g., Faulconbridge & Muzio, 2016), law firms’ M&A expertise is often
country-specific: Local experts are often regional or even local players, suggesting (target)
country-specific expertise as a key selection criterion.
Here, the mechanisms that enable a country expert to shorten closing time are particularly
relevant. First, familiarity with the legal and regulatory framework of a focal country implies the
ability to efficiently ensure compliance with all relevant regulations of this country (Gilson &
Black, 1996). Second, routines to apply efficient information verification techniques, which may
vary by country, allow for expediting legal due diligence, among others, because familiarity with
the specific context facilitates interpretation and an assessment whether and what additional
information needs to be collected from local information providers. Third, as suggested by
studies on cultural influences on negotiations, dealings with the seller and its legal counsel are
likely to be influenced by country context (e.g., Adair & Brett, 2005). Having routines in place to
carry out these negotiations should reduce completion time. Fourth, established relationships
with local authorities are an important form of social capital. They facilitate interpretation and
appraisal of regulatory requirements and preparation of necessary documents, thereby reducing
the time that the deal is being reviewed by the authorities. Overall, a law firm’s country-specific
legal expertise likely allows it to engage more effectively in ‘engineering’ a deal. Studies in
financial economics have indeed shown how bidders may benefit from hiring lawyers with
Selecting Legal Advisors in M&As 10
superior general and, to some extent, country-specific expertise during the final phases of
M&As, especially in terms of quicker closing (Karsten et al., 2014; Krishnan & Masulis, 2013).
Which of these lawyer characteristics—network relations or country-specific expertise—
is more important for an acquirer’s selection of a legal counsel; and which is a better predictor of
the counsel’s performance? For domestic M&As, we suggest that country-specific expertise is
less important in both respects: Acquirers generally face difficulties in assessing the quality of an
unknown lawyer, both ex-ante and ex-post (Coates, 2001). Heterogeneity among domestic law
firms in terms of their country-specific expertise in the domestic context is, by definition,
limited. In addition, acquirers are themselves familiar with the basic tenets of the legal
framework governing business operations in their home country. Therefore, we expect acquirers
to rely on their trusted counsel, as they stand to gain little from appointing an unfamiliar one; and
this choice will be in line with the drivers of lawyers’ performance. Overall, we expect:
Hypothesis 1a (H1a): In domestic M&As, network- and expertise-related attributes of a
law firm are positively associated with the likelihood of being selected as counsel, but
acquirers place relatively more weight on network-related criteria.
Hypothesis 1b (H1b): In domestic M&As, network- and expertise-related attributes of a
law firm are positively associated with its performance in terms of shorter closing time,
but network-related criteria have a relatively stronger impact on performance.
Acquirers’ Selection of Legal Advisors and Law Firm Performance in International M&As
First, selection: Do bidders transfer domestic advisor selection routines to the
international domain? Prior research offers contrary perspectives. Some scholars have argued
that acquirers in international deals are acutely aware of their unfamiliarity with the deal context,
Selecting Legal Advisors in M&As 11
including the institutional environment (e.g., Barkema & Schijven, 2008; Muehlfeld, Rao Sahib,
& van Witteloostuijn, 2012), suggesting an increased sensitivity to the need to adapt.
Yet, collecting information about competencies and reliability of another firm is costly
and time consuming (e.g., Stinchcombe, 1990). For legal services, this challenge is aggravated
by the fact that “legal advice often involves questions of judgment under conditions of
uncertainty that will persist even after a trial or negotiation or other legal event is completed”
(Coates, 2001: 1301). Hence, as acquirers face difficulties in assessing the quality of an unknown
(host country) lawyer vis-à-vis the proven qualities of a known advisor, they may transfer the
network-heavy weighting of selection criteria from the domestic to the international domain.
Research on social networks supports this view, pointing to “embeddedness as a solution to
relational hazards in market exchanges” (Lee, 2013: 1238). In the very situations that require
specific expertise to handle a complex task with uncertain outcomes, actors often resort to what
they perceive as the least risky option: their well-known, trusted partners (Lee, 2013; Mizruchi &
Stearns, 2001). This is not to say that acquirers necessarily select their counsel based on
irrational motives. Rather, the uncertainty they face in the nomination decision, combined with
the complexity and uncertainty of the cross-border deal itself, may make them highly risk averse,
‘nudging’ them away from the local expert, and towards the familiar counsel: “when trustworthy
partners are already available, searching for strangers is hard to be rationalized” (Chung et al.,
2000: 5). Indeed, Howard, Withers, Carnes, and Hillman (2016) found that greater firm-level
uncertainty led firms to reinforce alliance ties rather than engage in a broadening of ties. Thus:
Hypothesis 2a (H2a): In cross-border M&As, network- and expertise-related attributes of
a law firm are positively associated with the likelihood of being selected as counsel, but the
typical acquirer places relatively more weight on network-related criteria.
Selecting Legal Advisors in M&As 12
Second, performance: How could target country-specific expertise reduce closing time?
Essentially, we expect the same mechanisms to apply as proposed above: familiarity with the
legal framework, routines for information verification and legal due diligence, familiarity with
culture-specific negotiation styles and customs, established relationships with authorities.
However, compared to domestic M&As, acquirers are up against greater challenges in
attempting to complete international deals due to, among others, national differences in terms of
formal and informal institutions (Dikova et al., 2010; Weitzel & Berns, 2006), such as nation-
specific legal and tax systems and accounting practices (Very & Schweiger, 2001). This
complication affects, in particular, legal issues during the closing phase. Not only does the legal
service industry feature a high degree of national embeddedness (Faulconbridge & Muzio, 2016),
legal advisors additionally face the task of ensuring legal compliance in a highly fragmented
environment with increasingly complex national takeover regulations (White & Case, 2009).
Thus, the provision of legal services in M&As has only limited potential for developing any
generalized expertise (cf. Karsten et al., 2014). In addition, the limitations to transferability of
local expertise are exacerbated by the fact that many countries have installed widespread
impediments to the entry of foreign lawyers. Hence, a lawyer’s local expertise in a core country
of its operations constitutes a key asset in advising deals that involve this country. In sum, we
expect that a law firm’s target country-specific expertise allows it to engage more effectively in
“engineering” a deal including quicker legal due diligence, reduced time for drafting and revising
the contract and mark-ups, more efficient negotiations with the counter-party, and shorter time
that the deal is being reviewed by authorities. Given the eminent role of legal contingencies
during the public stage, we further maintain that in cross-border deals, target country expertise is
even more significant for advisor performance than a trustful relationship and reduced relational
Selecting Legal Advisors in M&As 13
frictions between bidder and advisor. Overall, we expect that the relative importance of network-
and expertise-related attributes is reversed, compared to domestic deals:
Hypothesis 2b (H2b): In cross-border M&As, network- and expertise-related attributes of
a law firm are positively associated with its performance in terms of shorter closing time,
but expertise-related criteria have a relatively stronger impact on performance.
In sum, while H1a-H2a fit into an (at least boundedly) rational capability-based logic of
partner selection, H2b breaks this logic: In cross-border deals, acquirers continue to rely on their
long-term partners, with ties initially established in domestic settings, but these embedded
lawyers are unlikely to be experts in the specific host country environment and vice versa (cf.
Mitsuhashi & Min, 2016). Social network research provides a possible rationale for such a
discrepancy between drivers of selection and of law firms’ performance in cross-border M&As:
Acquirers may favor familiar choices over the most capable ones, because they fall prey to a
behavioral bias arising from a dark side of embeddedness (e.g., Lee, 2013). Can experience
eliminate this discrepancy by stimulating learning? And, if so, could multiplicity of mental
models within the acquirer firm facilitate this learning process?
Types of Organizational Learning and Selection of Legal Advisors in International M&As
Organizational learning refers to an experience-driven change in an organization’s
knowledge, which includes both declarative and procedural knowledge, that is, facts as well as
skills and routines, thus embracing both cognitive and behavioral elements (e.g., Argote, 2013;
Fiol & Lyles, 1985). Routines represent collective, repetitive, and fairly stable patterns (Nelson
& Winter, 1982) that may relate to operational, administrative, and strategic activities, such as,
for example, procedures to select advisory services in M&As. By performing a task, an
organization accumulates experience, giving rise to learning.
Selecting Legal Advisors in M&As 14
Yet, there is more than one kind of learning. Different kinds of learning appear to co-
exist in organizations (e.g., Miller, 1996) and most scholars agree on the existence of at least two
types of qualitatively distinct learning (cf. Tosey, Visser, & Saunders, 2012). Tosey et al. (2012)
identify as the most common labels to capture this dichotomy: lower- and higher-level learning
(Fiol & Lyles, 1985), single- and double-loop learning (Argyris & Schön, 1978), exploitation
and exploration (Levinthal & March, 1993; March, 1991), incremental and radical (Miner &
Mezias, 1996), and adaptive and generative (Senge, 1990). Further, they suggest that a
“consensus seems to have been established that they refer to comparable learning processes and
outcomes” (Tosey et al., 2012: 292). Here, we use the terms lower- and higher-level learning.
Lower-level learning occurs within a given set of rules, norms, and frameworks, is based on
existing routines, maintains the core features of the relevant ‘theory-in-use’, and restricts “itself
to detecting and correcting errors within that given system of rules” (Fiol & Lyles, 1985: 808;
referring to Argyris & Schön, 1978). Higher-level learning implies changing underlying rules,
norms, and frameworks, and aims at developing “new cognitive frameworks within which to
make decisions” (Fiol & Lyles, 1985: 808). It requires “unlearning” of conceived knowledge and
reorientation of cognitive structures (Nystrom & Starbuck, 1984).
Acquirers’ Adaptation of Selection Criteria from Domestic to International Settings
Applied to our research setting, we characterize lower-level learning as attempts to
improve the selection of legal counsels in cross-border M&As without questioning the basic
selection rules, that is, for example, by adapting the weights attached to the two different kinds
of selection criteria transferred from the domestic context, but without categorically changing the
predominant importance of network-related attributes. Based on H1-H2, we argue, though, that
acquirers ultimately need to recognize that the relative significance of network- vis-à-vis
Selecting Legal Advisors in M&As 15
expertise-related attributes is reversed in cross-border M&As. This requires a switch from lower-
to higher-level learning.2 Are acquirers able to use their M&A experience to achieve this shift?
The value of experience for performance-enhancing learning crucially depends on the
extent to which it applies to a future activity (March, 1991), i.e., its transferability is bound by its
context-specificity. Prior studies have emphasized the limits of such transferability and, thus, of
the generalizability of prior experience (e.g., Haleblian & Finkelstein, 1999). Discriminating
between appropriate and inappropriate generalization appears to be particularly difficult in
complex corporate domains due, among others, to causal ambiguity (Argote, 2013). Firms may
need to accumulate a lot of experience before performance improves and, initially, more
experience may even reduce performance due to inappropriate generalization (e.g., Duijsters,
Heimeriks, Lokshin, Meijer, & Sabidussi, 2012; Haleblian & Finkelstein, 1999). Individuals as
well as organizations tend to suffer from resistance to adaptation, due, for example, to cognitive
biases (e.g., Samuelson & Zeckhauser, 1988) and greater uncertainty of exploring new
alternatives vis-à-vis refining existing ones (March, 1991). When the required learning entails a
shift from lower- to higher level, it is likely to be particularly challenging, as it requires ‘distant
search’, while the initial response to unsatisfactory outcomes tends to be ‘local search’ (Cyert &
March, 1963), which may result in further suboptimal outcomes (Barkema & Schijven, 2008).
When acquirers gain experience with selecting advisors, the repeated experience of prior
knowledge (about selection criteria) conflicting with to-be-learned information (antecedents to
advisor performance in cross-border deals), may give rise to the gradual emergence of novel,
alternative cognitive structures for how to select the most suitable legal advisor for a particular
type of deal—domestic or cross-border. Yet, it may take many M&As to achieve this. Thus, we
expect that acquirers’ initial learning will result, if anything, in greater weight of past ties as a
Selecting Legal Advisors in M&As 16
selection criterion, relative to (target) country expertise; and that they need to accumulate a lot of
experience before adequately appreciating a law firm’s (target) country expertise. The turning
point marks, as we argue, the sketched shift in selection routines.
Hypothesis 3 (H3): With growing cross-border M&A experience, acquirers first attach
more relative weight to network-related criteria and, after a turning point, increasingly
attach less weight to network-related relative to expertise-related criteria.
The Impact of Intra-Organizational Multiplicity of Mental Models
Prior research has identified three major types of moderators of the relationship between
experience and learning outcomes (Argote, 2013), that is, types of experience (e.g., direct vs.
indirect), attributes of the learning process (e.g., its mindfulness), and elements of the
organizational context (e.g., organizational culture). Adding to the third domain, we propose that
structural attributes of organizational cognition—in particular, multiplicity of mental models—
may foster a specific aspect of organizational learning, i.e., the ability to recognize, based on the
accumulation of experience, the need to shift from lower- to higher-level learning.
Mental models have initially been conceptualized at the individual level. Rouse and
Morris (1986: 360) characterized a mental model as a “mechanism whereby humans generate
descriptions of system purpose and form, explanations of system functioning and observed
system states, and predictions of future system states.” As such, mental models represent
“organized knowledge frameworks that allow individuals to describe, explain, and predict
behavior” (Lim & Klein, 2006: 404). Senge (1990: 163) characterized them as “images that limit
us to familiar ways of thinking and acting.” Similar to other related concepts (e.g., cognitive
map; for a review, see Walsh, 1995), a mental model thus constitutes a form of knowledge
structure or schema—a fundamental top-down information processing construct characterized by
Selecting Legal Advisors in M&As 17
Walsh (1995: 281) as a “mental template[s] that individuals impose on an information
environment to give it form and meaning”. Scholars have subsequently transferred the
knowledge structure construct to more aggregate levels and have, using a variety of labels,
discussed the existence of supra-individual knowledge structures at the team/group, organization,
and even industry level of analysis (cf. Walsh, 1995, for a review). For example, in the context of
product diversification, Prahalad and Bettis (1986: 491, 485) characterized dominant logic as a
“mind set or worldview or conceptualization of the business and the administrative tools […]
stored as a shared cognitive map”, which “consists of the mental maps developed through
experience in the core business”. Based on these studies, we define mental models at the level of
groups within the organization as collective organized knowledge frameworks that allow for
describing, explaining, and predicting behavior related to a specific information domain. For
example, we expect business units operating in different product markets to each be guided by
such a collective knowledge framework that allows the unit to describe, explain, and predict
behavior in its sector, related, for example, to industry conduct.3
It follows from the characterization of mental models as organized knowledge
frameworks that they have both content and structure (Walsh, 1995). Attributes relating to both
of these dimensions are thought to impact the consequences of these knowledge structures in
guiding future behavior. In terms of content, for example, knowledge structures that represent an
information environment build on experience in this domain, and the context in which this
experience was gained, constitutes an important boundary condition (e.g., Kaplan, 2008; Walsh,
1995). The emergence of knowledge structures appears to entail reflection upon and
interpretation of this experience, including the development of perceptional filters that affect
subsequent information acquisition and processing (e.g., Simons & Chabris, 1999). Here, we
Selecting Legal Advisors in M&As 18
focus solely on structure, though; and, unlike most prior studies concerned with structural
attributes, we consider the structure of a whole set of mental models rather than the structure of
an individual schema. Prior research on schema structure has analyzed differentiation, i.e., the
number of a knowledge structure’s internal dimensions, and integration, i.e., the extent of
interconnectedness of these dimensions (see Walsh, 1995, for a review). Using the term
‘multiplicity’ we consider differentiation at the level of a set of knowledge structures (e.g., the
number of group-level mental models associated with the product markets served by an
organization), a structural attribute that is also referred to as ‘richness’ in organizational ecology
(e.g., Boone, Wezel, & van Witteloostuijn, 2013). Richness captures a pure count of the number
of distinct types in a population or system. Accordingly, multiplicity of mental models refers to a
count of the number of distinct mental models relating to a given information domain within an
organization. Thereby, our definition of multiplicity is both related to and yet distinct from
diversity as commonly conceptualized in management research (see Harrison & Klein, 2007, for
a review). Harrison and Klein (2006) outlined three major ways in which the diversity construct
is being used, that is, diversity as separation, disparity, and variety. The latter is closest to our
definition of multiplicity. Yet, diversity as variety typically takes into account both richness and
abundance or relative distribution of these types (i.e. the proportion of unit members who belong
to a particular type). Thus, multiplicity is a simpler concept, without specific assumptions on
more complex distributional characteristics.
In terms of influencing an acquirers’ ability to shift from lower- to higher level learning,
we argue that multiplicity of mental models leverages the organization-level “ability to reflect
on, understand, and control” (Schraw & Dennison, 1994: 460) the entity’s learning, in particular,
by facilitating the formulation of strategies for considering alternative cognitive mechanisms and
Selecting Legal Advisors in M&As 19
for deliberately selecting from a set of available mechanisms those that are to be applied in a
specific situation (Flavell, 1987). Thus, we propose that multiplicity enables firms to better
utilize their (domain-specific) experience as input for their learning by facilitating more accurate
reflection on the generalizability of this experience: First, multiplicity of mental models relating
to a certain domain may have task- or domain-specific effects—akin to potentially positive
effects from diversity of experience (e.g., Haunschild & Sullivan, 2002; Nadolska & Barkema,
2014)—by making available a range of cognitive structures for describing and dealing with a
specific type of task (e.g., selecting lawyers). By allowing for consideration of a larger range of
perspectives, qualitatively better decisions may become possible (e.g., Bell, Villado, Lukasik,
Belau, & Briggs, 2011; Dahlin, Weingart, & Hinds, 2005), with positive effects of heterogeneity
potentially resulting from superior information processing related to range, depth, and integration
of information use.
Second, resting on our conceptualization of multiplicity of mental models as a content-
independent, structural attribute, we propose that it may additionally stimulate a generalized
awareness effect independent of the specific task or domain. Each mental model rests on a set of
assumptions, rules, and norms. Greater multiplicity of mental models illustrates the possibility to
view the world through a larger number of sets of rules. By raising awareness of the existence of
many distinct worldviews, such multiplicity may facilitate a specific aspect of organizational
learning, that is, recognition of the need to switch from lower- to higher-level learning that
implies changing the underlying rules. Thereby, multiplicity of mental models could allow for
more accurate reflection on the categorization, transferability, and generalizability of
experience—processes that are fraught with difficulties, as prior studies have emphasized (e.g.,
Haleblian & Finkelstein, 1999). Based on the assumption that the co-existence of a multitude of
Selecting Legal Advisors in M&As 20
worldviews in a domain is widely known throughout the organization, acquirers with greater
multiplicity of mental models should thus require less experience to move from lower- to higher-
level learning. Hence, our arguments as to this generalized effect imply that we expect the
beneficial effects of multiplicity to apply beyond a specific task, ranging from closely related
domains to fairly distant and even unrelated ones (for a similar argument, see Powell & Rhee,
2016). What matters are not actual similarities between two tasks, but demonstration of the—at
least hypothetical—possibility to ‘do things differently’. Thus, while experience represents a
major building block for knowledge structures such as mental models (Walsh, 1995), knowledge
about the co-existence of worldviews related to a specific domain (e.g., product markets
associated with business units) may permeate the organization in a way that knowledge about
specific experiences in this domain could not. Thereby, the proposed generalized effect differs
from concepts such as knowledge transfer between units, which often suffers from stickiness that
impedes the reuse of practices in other subunits (Szulanski, 1996). Overall, we expect:
Hypothesis 4 (H4): Multiplicity of mental models is positively associated with an
acquirer’s experiential learning: Acquirers with greater multiplicity of mental models
require less cumulative cross-border M&A experience to reach the turning point beyond
which they attach relatively less weight to network- vis-à-vis expertise-related criteria.
METHODOLOGY
Data and Sample
We retrieved data from the Thomson Reuters’ Financials SDC M&A database
(henceforth: Thomson). It records all publicly announced M&As worldwide, including details on
all legal counsels involved in the transactions. Our initial Sample A consists of 153,622 domestic
and international M&A attempts announced in 1994-2010. We used the available information on
Selecting Legal Advisors in M&As 21
acquirers and counsels to construct our main variables, whereby our definition of acquirer and
counsel coincides with the variables ‘acquirer ultimate parent’ and ‘legal advisor’ in Thomson.
We then reduced Sample A to a smaller set (5,594 domestic; 5,917 cross-border M&As) for our
hypotheses tests. This Sample B includes acquirers and deals (a) that are neither self-tenders nor
minority deals, (b) that have at least one law firm on board, (c) with acquirer and target primarily
active in the non-financial non-legal service industries, (d) where the acquirer attempted at least
one M&A in the four years prior to the focal deal, and (e) where all relevant variables are filled.4
Dependent Variables
For testing H1b and H2b, we generated for each M&A in Sample B the integer variable
duration of the closing phase. It measures the difference between the dates of initial public
announcement and of unconditional completion for every merger officially consummated before
January 11th, 2015. Because durations range from 1 to 3213 days, with a median of 53 days, we
included their natural logarithm, thereby alleviating concerns that the results might be driven by
a small number of deals with a very long closing phase. However, the correct unit of analysis for
testing H1b and H2b is not a merger, but an advisor-acquirer dyad. In order to avoid the
statistical interdependency between the considerable number of M&As in Sample B (34%),
where the acquirer employed a team of two or more advisors, we focused on the ‘lead advisor’,
i.e., the largest lawyer in terms of law firm worldwide experience (variable introduced below).5
For testing H1a, H2a, H3, and H4, we constructed for each M&A in Sample B a ‘risk set’
of potential legal counsels. Within this set, we examined the probability that a given law firm
was selected by the focal acquirer, measured as the likelihood of selection, which is 1 for every
law firm that was appointed (0 otherwise). The risk sets include, next to the actually appointed
counsels, also lawyers that could have potentially been selected. As there are many firms that are
Selecting Legal Advisors in M&As 22
obviously implausible candidates, e.g., a German lawyer in a U.S.-Japan deal, we constructed
sets of ‘shortlisted’ candidates, that had counseled in at least one other deal in the year of the
focal M&A’s announcement, and were among the top 5 percent of firms in terms of at least one
of the four network- and expertise-based law firm attributes described below. A typical risk set in
our analysis then consists of, on average, 104 law firms per deal. Multiplied by the 11,511 deals
in Sample B, the total risk set (Sample C) comprises 1,160,232 acquirer-law firm dyads.
Independent Variables
Advisor-client relationship. We measured the intensity of the law firm-acquirer
relationship by prior interactions between them. We counted for each law firm-acquirer dyad in
Samples B and C the number of (prior) advisor-client ties as recorded in Thomson and interpret
this, in line with Gulati (1995), as a measure of inter-organizational familiarity. Further, we
divided this count measure by the total number of law firms an acquirer had previously worked
with to obtain the percentage variable advisor-client tie strength. Following network literature
(e.g., Jensen, 2003; Mizruchi & Stearns, 2001), this measure serves as an indicator of inter-
organizational trust, as it captures the degree of ‘exclusivity’ of a relationship. The data—as for
all law firm-related variables—was retrieved from Sample A. We counted all matches of the
same name pair in the transactions prior to a focal M&A that precede the deal by no more than
four years. Prior studies on organizational networks have often used such a moving time window
(e.g., Gulati & Gargiulo, 1999; Lee, 2013). For the legal counsel market, it is further justified by
high turnover of personnel, especially in the elite of the profession (Heinz et al., 2001).
Law firm country-specific expertise. We operationalized country-specific expertise by
law firm acquirer country expertise and law firm target country expertise (henceforth acquirer
country expertise and target country expertise). Each variable counts a firm’s prior M&A
Selecting Legal Advisors in M&As 23
advisory appointments in the home country of the focal acquirer and the focal target,
respectively. Using transaction history as expertise measure has several advantages over
alternatives; in particular, as it captures, next to knowledge embedded in the firm, also its
business experience. Uzzi and Lancaster (2004), for example, measure legal expertise of a law
firm directly by weighting the law school degrees of lawyers in each of a firm’s (foreign) offices.
Yet, as expertise is rewarded with more offers, the number of past appointments is a viable proxy
for these more direct dimensions of legal expertise (Lazega, 2001). At the same time, it portrays
a more accurate picture of the service quality, as the transaction history also captures experience
and routines developed through actual practice of the law. We calculated acquirer and target
country expertise as the percentage shares of the total number of a law firm’s past appointments
in the four years prior to a deal. A value of 100% (0%) reflects high (low) expertise in the legal
framework of a country. In this way, the variables capture those elements of service quality that
are difficult to transfer across national contexts. Only very few large law firms are truly
international: Linklaters (U.K.) and Clifford Chance (U.S.), for example, advised foreign clients
in 62%, and 75% of their deals during our sampling period, respectively. The vast majority of
firms, even among the top tier, focus almost entirely on a domestic clientele. By controlling for
overall transaction history, this percentage measure thus reflects the high degree of geographical
specialization in the industry.
Acquirer cross-border M&A experience. Acquirer cross-border M&A experience counts,
for every acquirer in Sample B, all prior cross-border M&As attempted by the firm in the four
years preceding the focal deal. For testing H3, the variable is included in the legal advisor
selection models, where it is interacted with the law firm variables described above. As learning
Selecting Legal Advisors in M&As 24
might be non-linear (e.g., Haleblian & Finkelstein, 1999), we also included the square of cross-
border M&A experience in the selection models.
Acquirers’ multiplicity of mental models. Measuring multiplicity of mental models is
challenging, especially in large-scale research settings, which allow only for indirect assessment
(for direct methods with smaller samples, see, e.g., Kaplan, 2008; Markoczy, 2001). Thus, we
used several measures, which vary along two dimensions. First, in terms of level of analysis, we
constructed three measures at the organizational level and one at the level of the top management
team (TMT). Second, regarding distance from the object of learning, we included several
corporate domains, ranging from more to less distant from selecting legal advisors in M&As.
At the organizational level, we assessed multiplicity of mental models (mmm), first, in
the domain of antitrust systems (e.g., Glendening, Khurana, & Wang, 2016). We classified all
countries in Sample A according to the Wiki project “Antitrust around the World” (Hylton &
Deng, 2007). This Wiki identifies core dimensions of antitrust systems and cardinally ranks
countries’ current and historic systems in terms of the scrutiny they exercise in each of them. We
used the overall antitrust system score. As the overall country scores are clustered between 17
and 25 points, we grouped countries by score percentiles to obtain a more discriminatory
classification. Our theory emphasizes the impact of the simple prevalence of a range of distinct
mental models rather than trying to weight them according to their frequency or relatedness—as
is often done in studies focusing on diversity of experience (e.g., Haunschild & Sullivan, 2002;
Powell & Rhee, 2016). In contrast to such studies, we therefore constructed a simple count
measure (mmm antitrust systems) of the number of different antitrust score percentiles an
acquirer had been exposed to during the four years prior to a focal deal (similarly, see, e.g.,
Dahlin et al., 2005).
Selecting Legal Advisors in M&As 25
Second, as M&As are frequently subject to political manoeuvring (Serdar Dinc & Erel,
2013), we measured multiplicity of mental models in the domain of political systems (mmm
political systems). We classified the countries in Sample A according to the ‘political stability’
indicator of the Worldbank Worldwide Governance Indicators database. The index is a cardinal
number ranging from -2.5 to 2.5 and is available as of 1996. Similar to the mmm antitrust
systems measure, we counted for each acquirer the number of different political systems’ score
percentiles that the firm had been exposed to in the four years prior to the focal deal, which
means that the variable is available for deals announced in 2000-2010. Both of these measures
have the advantage that they depict the focal structural attribute of an organization’s mental
models (i.e., their multiplicity) in a domain that is closely related to the object of learning, i.e.,
selecting legal counsels in cross-border M&As. Yet, for this very reason, it is difficult to
disentangle their effect from the impact of experience (in terms of both volume and diversity) in
these domains. In fact, experience constitutes a major building block of the emergence and
development of any knowledge structure and thus, also a key driver of multiplicity of mental
models. Therefore, in order to test H4, we also sought to assess mental models in less related
corporate domains that bear no immediate relation with the selection of legal counsels in M&As.
Thus, third, we considered the domain of product markets. Prior research has argued that
product markets or industry sectors are associated with distinct, industry-specific shared concepts
of business conduct (Spender, 1987), which may be associated with distinct mental models (e.g.,
Ginsberg, 1989, on the impact of diversification on the content of managers’ mental models).
Various ways exist to assess the degree to which firms have operations in more than product
market, such as the Herfindahl index (e.g., Wan, Hoskisson, Short, & Yiu, 2011). Again, in line
with our theory, we instead used a simple count of the total number of distinct 4-digit SIC codes
Selecting Legal Advisors in M&As 26
in which an acquirer was active at the time of announcing the focal deal. Thomson records for
both acquirer and target in every merger all their industry activities as measured by SIC codes.
Since our unit of analysis is the acquirer’s ultimate parent, we aggregated the subsidiary SIC
codes and counted for every ultimate parent in Sample B the number of unique SIC codes of all
affiliations with the parent at the time of announcing the focal M&A and in the prior four years.
The result is the time-varying variable acquirer mmm product markets.
Fourth, finally, we included a measure of multiplicity related to a firm’s TMT. We
measured the number of nationalities represented among the top decision-makers, divided by the
size of the TMT. Upper Echelon (UE) studies (e.g., Bromiley & Rau, 2016; Hambrick,
Davidson, Snell, & Snow, 1998; Nielsen & Nielsen, 2013) and team studies in general (e.g.,
Dahlin et al., 2005) suggest that nationality/national origin may be particularly strongly related to
the worldviews of decision-makers and the mental models they hold (see, related, Marano,
Arregle, Hitt, Spadafora, & van Essen, 2016; although this view is not uncontested, e.g.,
Markoczy, 1997; for details, see online appendix 1). Note that our operationalization differs from
the way in which UE and team studies usually assess team composition, often based on the Blau
index (e.g., Carpenter & Frederickson, 2001; in relation to TMT diversity and acquisition
learning, e.g., Nadolska & Barkema, 2014), as we are interested purely in how many distinct
mental models are present in a TMT, rather than in any group dynamics. As TMT data are not
recorded in Thomson, we collected this information from archival sources. This, though,
required us to focus, in line with extant UE studies (e.g., Carpenter & Frederickson, 2001), on a
smaller sample of deals and associated acquirers. In order to keep the data collection tractable,
while maximizing the chances of finding the required TMT information, including as well non-
U.S. firms, we zoomed in on the year 2010 and on the largest 50 acquirers that appeared at least
Selecting Legal Advisors in M&As 27
once during that year. Following prior studies (e.g., Carpenter & Frederickson, 2001), we
operationalized the TMT as the top two tiers of management: CEO, members of the Executive
Board (e.g., chief financial officer) and the next-highest management tier including other senior
managers listed as executive officers. We obtained information from firms’ annual reports, SEC
10-K Filings, and, if otherwise not available, from company websites and public sources such as
directorstats.co.uk. Thereby, we were able to acquire complete information for 30 firms involved
as acquirers in 65 cross-border M&As from among our sample in 2010.
Control Variables
In terms of lawyer attributes, based on the close link between service quality and market
rewards (Lazega, 2001), law firm worldwide experience captures general quality differences
between firms. In a service industry such as the legal sector, it also proxies for firm size. Number
of advisors controls for the influence of the size of the advisory team (Hunter & Jagtiani, 2003).
Considering acquirer attributes, as M&A experience has been shown to also have a direct,
potentially non-linear, influence on pre-completion outcomes (e.g., Dikova et al., 2010), we
included an acquirer’s cross-border M&A experience and the squared experience as control
variables in the outcome analyses. For the same reason, we also included the mental model
variables and acquirer country experience and target country experience in these analyses. We
constructed the latter variables in the same way as the corresponding law firm variables.
Acquirer network size counts the number of different law firms a bidder has previously worked
with. Together with advisor-client tie strength, this variable defines an acquirer’s ‘market
interface’ (Baker, 1990) with the legal industry. Public acquirer indicates whether an acquirer is
publicly listed (1/0 otherwise) and acquirer size measures the total asset value of an acquirer and
its subsidiaries (in U.S.$). In terms of transaction attributes, based on prior research (e.g., Wong
Selecting Legal Advisors in M&As 28
& O’Sullivan, 2001), we included public target (1/0 otherwise); deal value (in U.S.$); number of
bidders; the percentage variable stock payment measuring the acquirer’s intended share of stock
payments; and tender offer to identify deals where the acquirer approaches the target’s
shareholders directly (1/0 otherwise). We controlled for timing of a deal by including year
dummies. Based on La Porta, Lopez-de-Silanes, and Shleifer (2008), we controlled for regulatory
hurdles by adding a common law origin dummy if the target country has a common law origin,
i.e. a more merger friendly control regime (Rossi & Volpin, 2004) (1/0 otherwise). Finally, same
legal origin indicates acquirer-target country pairs with the same legal origin (1/0 otherwise).
Estimation Techniques
The major issue with our simultaneous estimation of selection criteria and outcomes is
that the law firms in Sample B are not randomly allocated to deals. Instead, they are selected by
the decision of an acquirer who chooses, as hypothesized, only the most suitable candidates from
Sample C. This may confound the true effect of a law firm’s network- and expertise-based
attributes in the outcome analysis (Cameron & Trivedi, 2005). Even though this problem is
partially alleviated by including various control variables, which together account for the likely
fact that a more difficult case calls for a more competent lawyer, several dimensions of
transaction complexity that we do not observe are likely to remain. To account for these
unobserved dimensions, we estimated Heckman two-stage models (Heckman, 1979) with the
likelihood of selection as the dependent variable in the first stage and duration in the second
stage of our regressions. These models require some regressors in the selection equation that are
not included in the outcome equation, but none of the aforementioned variables qualifies as such
an exclusion restriction, as we constructed all of them based on the assumption that they might
affect outcomes. We therefore followed recent research in financial economics (e.g., Acharya,
Selecting Legal Advisors in M&As 29
Davydenko, & Strebulaev, 2012) and generated several auxiliary interaction terms from the
available law firm variables that capture the notion of ‘herd behavior’: An acquirer is likely to
put more weight on an attribute as selection criterion, if more other acquirers rely on the same
criterion. We opted for law firm worldwide experience, interacted it with the variable’s average
among all lawyers appointed as legal counsel in the same country and same year as the focal
deal, and included the resulting interaction term in the selection equations.
RESULTS
Summary Statistics
Table 1 presents means, SDs, and Pearson correlation coefficients of all variables in
Sample B (see online appendix 2 for the corresponding table for the larger Sample C.) As the
variable range of many of the count measures is substantial, while their means and medians are
comparatively small, we computed the natural logarithms to ensure that our regression results are
not driven by a few firms with very large counts. Correlations between the mmm variables vary
between -.06 and .78, supporting the notion that they indeed describe knowledge structures in
different domains (Walsh, 1995). Also, the correlations between (a) the mmm variables and (b)
acquirer cross-border M&A experience and acquirer size, respectively, are—with a maximum of
P.C.<.81—in the range of values typical of experiential learning studies (e.g., Haunschild &
Sullivan, 2002).
--------------------------------------
Insert Table 1 about here
--------------------------------------
Acquirers’ Criteria for Selecting Legal Advisors
Selecting Legal Advisors in M&As 30
Table 2 presents the results of testing H1a and H2a. Model 1 (2) investigates the selection
criteria that acquirers apply in their choice of a legal counsel for a domestic (cross-border) deal.
--------------------------------------
Insert Table 2 about here
--------------------------------------
H1a suggested that an advisor would more likely be selected for a domestic deal, if it had
more country expertise and closer connections with the acquirer. Indeed, advisor-client ties,
advisor-client tie strength, and acquirer country expertise are all significant and positive in
Model 1. We also posited that network-related criteria would have a stronger impact on
selection. As both advisor-client tie strength and acquirer country expertise are measured in
percentage points, we tested this formally by comparing their unweighted effect sizes. In
particular, as suggested by Wiersema and Bowen (2009), we compared their marginal effects.
The results showed that advisor-client tie strength alone is a more important selection criterion
than acquirer country expertise over the entire range of observations, i.e., acquirer-law firm
dyads in Sample C (p<.021). Thus, H1a is supported. H2a suggested that acquirers would
transfer selection rules from domestic to cross-border M&As using basically the same relative
weight of criteria, with prior ties being predominant. In Model 2, the coefficients of all four law
firm attributes are significant and similar in size to the ones of Model 1. Also, a test on the
relative size of marginal effects of advisor-client tie strength vs. the country expertise variables
shows that even tie strength alone is more important for selection than the two expertise-related
variables together (p<.001). H2a is therefore supported.
Selecting Legal Advisors in M&As 31
Determinants of Pre-Completion Outcomes
First, we check in how far sample selection is an issue and in how far our exclusion
restriction is a valid instrument. Our exclusion restriction, the ‘herd variable’, is included in the
selection equations in Models 1 and 2 (Table 2). The coefficient is highly significant (p<.001),
suggesting that acquirers indeed seem to follow other acquirers in their selection criteria,
supporting our case of a valid instrument. The final lines of Models 7-10 (Table 3) present the
coefficients of the Inverse Mills ratio constructed from the selection equations and included in
the corresponding outcome equations. The ratio is only significant in the control variable
specification for domestic deals (Model 7, p=.002). It is insignificant in all other models
(p>.438), suggesting that sample selection is not a major issue for the analysis, in particular not
for the cross-border models. Because Heckman selection models are prone to produce
inefficiently high standard errors, Models 8-10 present, as recommended (Cameron & Trivedi,
2005: 552), coefficients and p-values of a standard OLS regression.
--------------------------------------
Insert Table 3 about here
--------------------------------------
H1b predicted that a lawyer’s country-specific expertise and strength of links with the
acquirer would reduce closing time in domestic M&As. According to Model 9, only advisor-
client tie strength has the hypothesized negative effect (p<.001). Acquirer country expertise has
the predicted sign but is insignificant; advisor-client ties has no meaningful effect. We also
hypothesized that prior relations would contribute relatively more to completion than country
expertise. This is confirmed in a Wald test on the relative strength of advisor-client tie strength
against acquirer country expertise (p=.016). Thus, H1b is supported for the effect of advisor-
Selecting Legal Advisors in M&As 32
client tie strength. H2b predicted a significant role of expertise and prior ties also for the
completion of cross-border deals, but, in contrast to H1b, with country expertise as the pre-
dominant factor. The first part of H2b is partly supported; the second part is fully supported:
Only acquirer country expertise and target country expertise have the predicted negative effects
on duration, whereby the latter shows a stronger effect (Model 10). The two relational variables
are insignificant. A Wald test on the relative strength of target country expertise vs. advisor-
client tie strength shows that the difference in effect sizes is statistically significant (p=.008).
H2b is supported, especially for a counsel’s target country expertise in cross-border deals.
In terms of economic impact, raising target country expertise from the average value of
0.21 in Sample C to the maximum of 1.0 reduces the closing phase of an international deal by
21%. At an average duration of 80 days, this corresponds to a reduction of ~17 days. The effect
of advisor-client tie strength in domestic deals is equally sizable: Duration decreases by 26% or
21 days. Both effects are in the range of values found for established determinants of pre-
completion outcomes (Wong & O’Sullivan, 2001). They are larger than, for example, the
difference between a private and a tender offer for an attempted cross-border deal (15 days) and
almost match the difference between an all-cash and a stock financed deal (26 days).
Testing for Learning and the Impact of an Acquirer’s Multiplicity of Mental Models
H3 postulated that the discrepancy between selection criteria and drivers of law firm
performance in cross-border deals would be curvilinear in bidders’ cross-border M&A
experience, with a turning point beyond which acquirers would predominantly consider country
expertise in their selection. This curvilinear effect is tested in Model 3 (Table 2) by means of
four interaction terms between acquirer cross-border M&A experience, the square of experience,
and the two key variables of our outcome models: advisor-client tie strength and target country
Selecting Legal Advisors in M&As 33
expertise. If such a curvilinear effect were to exist, we would expect that the moderating effect of
bidder’s cross-border M&A experience on advisor-client tie strength as a selection criterion
(EL1) was significantly larger than the moderating effect of bidder’s experience on target
country expertise (EL2). Also, we would expect the moderating effect of the square of bidder
experience on advisor-client tie strength (EL3) to be lower than the moderating effect of the
square of experience on target country expertise (EL4). Both conjectures are supported (Table
2). They can, moreover, be corroborated in a comparison of the true moderating effects, as
recommended by Wiersema and Bowen (2009): In more than 99% of our acquirer-lawyer dyads,
the moderating effects (ME) of experience are statistically significant (p<.05 across both
comparisons) and satisfy as predicted in H3: ME(EL1)-ME(EL2)>0 and ME(EL3)-ME(EL4)<0.
Hence, with sufficient experience, acquirers may overcome the initial bias in their international
selection decisions. Another question is whether the typical acquirer will ever accumulate
enough experience to reach the turning point. The corresponding marginal effects plot shows that
a transaction history of at least 13 cross-border deals is necessary to reach this point (see online
appendix 4, Figure A1). This is the privilege of 12.5% of the most frequent acquirers: For the
majority of firms in our sample, experience alone is thus insufficient.
Models 4-6 (Table 2) test whether bidders with a greater number of mental models reach
the turning point earlier, requiring less experience (H4). The results largely support this
hypothesis. First, the positive difference between the linear parts of the experiential learning
interactions, EL1 and EL2, prevails in Models 4-6, suggesting that, at low levels of experience,
acquirers lean increasingly more on advisor-client tie strength for selection. At high levels of
experience, in turn, acquirers put more weight on target country expertise for selection,
especially if they also exhibit greater multiplicity of mental models. This is suggested by the
Selecting Legal Advisors in M&As 34
positive difference between the coefficients on the mental model interaction terms (MM2-
MM1>0 in all three models). To confirm this formally, we also computed the true second-order
moderating effects (2ME) of our mmm variables. As the code is not by default implemented in
the statistical software package, we predicted the moderating effects manually based on the
STATA source code printed in the appendix of Wiersema and Bouwens (2009: 690). Online
appendix 5 presents our code and Table A4 an overview of the results. Online appendix 8
presents additional robustness checks. To summarize, the moderating effects of all three mmm
variables satisfy, as predicted in H4: 2ME(MM1)-2ME(MM2) < 0 for at least 90% of the
acquirer-law firm dyads in Sample C. Thus, the turning point seems contingent on an acquirer’s
multiplicity of mental models, i.e. multiplicity shifts down the critical number of M&As required
before an acquirer puts more weight on a lawyer’s target country expertise. The marginal effects
plots of Figure 1 illustrate this, too: The typical bidder with low multiplicity never reaches the
turning point, regardless of the amount of experience it accumulates as observed in our data. In
contrast, almost all acquirers with a high level of multiplicity reach the turning point (see Table
A5 in online appendix 6 for details). Finally, we assessed H4 in a sub-sample of deals announced
in 2010. We computed correlations between the number of TMT nationalities of the acquirers
and attributes of their counsels (see also online appendix 7, Table A6). In support of H4, TMT
nationalities is positively correlated with, and only with, target country expertise (P.C.=.29,
p<.001).
DISCUSSION
In the context of acquirers’ selection of legal advisors in M&As, this study, first,
examined whether firms transfer specific routines and rules from one context (domestic) to
another (cross-border); further, we analyzed whether such a transfer might result in an (initially)
Selecting Legal Advisors in M&As 35
suboptimal match (in terms of legal counsels’ performance) between the transferred routines and
the novel context. On this basis, the study investigated whether accumulation of M&A
experience would allow acquirers to overcome any such mismatch by revising their selection
criteria in cross-border deals. Such rebalancing, we argued, would require a categorical reversal
of the relative importance attached to the two types of law firm attributes studied—country
expertise vis-à-vis network relations. We suggested that acquirers’ experiential learning would
need to involve a shift from lower-level learning, which occurs within a given set of rules (Fiol
& Lyles, 1985), to higher-level learning that would entail adjusting these rules themselves.
Finally, this study examined whether acquirers featuring a greater number of distinct mental
models would enjoy advantages in terms of a specific aspect of their organizational learning, i.e.,
their ability to recognize earlier the need to switch to higher-level learning.
The empirical results of this study indicate that selecting lawyers for domestic and
international mergers indeed calls for distinct selection rules: In domestic deals, bidders
predominantly select their legal counsels based on prior network relations, which also dominate
as determinants of lawyers’ performance towards shortening time to completion. In cross-border
deals, acquirers tend to rely as well (at least initially) on lawyers with whom they share prior
business relationships. Yet, these law firms are less likely to be experts in the specific target
country environment. In line with extant social network studies, we thus find evidence in this
novel empirical context of a dark side of embeddedness, i.e., acquirers (at least initially) pay
insufficient attention to better matching partners outside their network. Acquirers appear to learn,
though, when they accumulate M&A experience. After counter-productive lower-level learning
at first, they reach a turning point (after a large number of deals) at which they categorically
reverse the relative weights of network- and expertise criteria. Finally, acquirers with a greater
Selecting Legal Advisors in M&As 36
number of mental models needed less experience before reaching this turning point. Overall, we
believe that our results have important implications for theory and practice.
Organizational learning research. First, conceptually and empirically, this study builds
on and extends research into components of the organizational context that are conducive to
organizational learning (e.g., Argote, 2013; Argote & Miron-Spektor, 2011; Fiol & Lyles, 1985).
In particular, it adds to research on organizational learning by investigating multiplicity of
mental models as a complementary facilitator of a specific aspect of firms’ experiential learning:
the ability to recognize the need to switch from lower- to higher-level learning. The ability to
identify when to engage in which type of learning has been highlighted as an important element
of organizations’ learning already by Fiol and Lyles (1985). To the extent that it is closely
related to reflection on the underlying learning processes itself, our results also speak to studies
that seek to establish foundations of a third type of learning such as ‘meta-learning’ (Visser,
2007) drawing, in particular, on Argyris and Schön (1996) (for overviews, see, Visser, 2007;
Tosey et al., 2012). The shift from lower- to higher-level learning that is central to this study
adds an outcome-oriented element to this concept. By empirically documenting such a shift, and
identifying factors conducive to it, we also contribute empirically to this literature, which is
largely conceptual in nature or rests on anecdotal evidence. Further, with respect to H4, the
notion that a firm’s level of diversification might be related to “skills for higher-level learning”
(Fiol & Lyles, 1985: 811) has been voiced earlier, although the specific mechanisms that could
give rise to such an association have mostly not been spelled out explicitly (e.g., Fiol & Lyles,
1985; related, Miller, 1996). For example, Fiol and Lyles (1985: 811) concluded their conceptual
study with questions for future research, one of them being whether “diversified firms have
better skills for higher-level learning than do single business firms”. Conceptually, we propose a
Selecting Legal Advisors in M&As 37
possible mechanism—multiplicity of mental models—that could give rise to such an effect and
empirically test the resulting prediction for a specific aspect of learning, i.e. recognition of the
need to shift to higher-level learning.
Our findings may also hold value for research at the interface of organizational learning
and diversification (e.g., Dujisters et al., 2012; Pennings, Barkema, & Douma, 1994). While the
dominant view in the literature on diversification is skeptical regarding its implications for firm
value (e.g., Lang & Stulz 1994; Wan et al., 2011), recent studies cautiously point to potential, yet
contingent benefits of diversification—for some firms, in certain contexts, and in relation to
particular tasks (e.g., Mackey, Barney, & Dotson, 2017; Tate & Yang, 2015). Our results add a
specific aspect of learning to the list of corporate activities that could be positively affected by
diversification, including as well a suggestion for a possible mechanism underlying this effect.
Social network research. This study adds to social network studies by focusing on how
experiential learning may provide a safeguard against the ‘dark side’ of strong network ties that
may lead organizations to pay insufficient attention to better matching partners outside their
network (e.g., Lee, 2013; Mitsuhashi & Min, 2016). Our results support this view by
documenting the contingent value of network relationships in a novel empirical context (e.g.,
Gulati, 1995; Yang et al., 2011). In addition, we show that firms appear to face significant
difficulties in overcoming the risks of over-embeddedness, as they need to accumulate
substantial experience before starting to pay due attention to local experts outside their networks.
Factors such as causal ambiguity and infrequency of experiences complicate organizational
learning in complex corporate domains (Argote, 2013). In the present context, a complementary
influence may impede bidders’ learning: strategic behavior of familiar partners. Past ties may
become a source of opportunistic exploitation, as the creation and presence of trust generates
Selecting Legal Advisors in M&As 38
precisely those conditions under which fraudulent behavior yields the highest payoffs
(Granovetter, 1985). As familiar counsels gain from exclusive relationships, they might,
strategically or unintentionally, obstruct acquirers’ learning by striving to exclude more
competent (local) experts from the counseling team.
M&A research. This study adds to literature on M&A completion (e.g., Chakrabarti &
Mitchell, 2016; Dikova et al., 2010; Muehlfeld et al., 2012; Zhou et al., 2016) by pointing to
legal advisors as a complementary influence. The explanatory power of the results, compared
with prior studies, implies that this type of factor—external advisors whose relevance likely
varies depending on the specific merger phase—is important to include in future studies.
Implications for practice. First, the results shed light on the value of experience for
learning how to overcome the dark side of embeddedness. They suggest that learning might be
impeded by external parties who strive to exclude competent (local) experts from their alliances
for strategic reasons—and acquirers should be aware of this. Second, positive effects for
organizational learning may result from the presence of a whole range of mental models
available in an organization. These findings, thus, point into a similar direction as other studies
such as Tate and Yang (2015) who argue that firms may reap productivity benefits from
redeploying labor internally to alternative uses—a type of human capital investment that likely
contributes to the diffusion of knowledge about multiplicity of mental models throughout the
organization. In a similar vein, Powell and Rhee (2016) emphasized the value of variance
(though in experience), among others due to greater caution with which decision-makers
interpret past experience. Embracing a healthy dose of skepticism as to the generalizability of an
organization’s own experience and keeping an open mind for alternative worldviews may enable
firms to more easily engage in rule-changing higher-order learning, if needed.
Selecting Legal Advisors in M&As 39
LIMITATIONS AND DIRECTIONS FOR FUTURE RESEARCH
As any study, this study has its limitations, which could be addressed in future research. A
first limitation relates to internal validity. Future research might complement our dependent
variable with other measures, such as, for example, completion time including search time
(related, Uzzi & Lancaster, 2004). Yet, the relevance of different types of external advisors (e.g.,
lawyers, investment banks; see, e.g., Hayward, 2003) varies across the M&A process. Outcome
variables and scope of influence of the type of advisor should be aligned. For example, while
legal advisors impact the speed of closing, they are not usually heavily involved in bargaining
about premiums. Still, future research should probe whether our conclusions extend to other
combinations of advisor-outcome variables. A second limitation pertains to our measures of
organizational learning and multiplicity of mental models, neither of which is directly observed.
It is standard procedure in much of the learning literature to use counts of cumulative experience
as input measures (for an overview, Argote, 2013). Still, future research should probe the
robustness of our results using alternative measures and settings, as our data do not allow us to
unambiguously identify the factors that drive the initial discrepancy between selection criteria
and antecedents of lawyers’ performance in international deals, and the sluggishness of
acquirers’ adaptation. Consequently, we cannot fully rule out alternative explanations: For
example, even if a familiar advisor cannot provide the same high service quality as a local
expert, the long-term advisor might charge lower fees (Uzzi & Lancaster, 2004) or be more
trustworthy in terms of the variability of outcomes (cf. Baum et al., 2005).
Further, operationalizing multiplicity of mental models and disentangling the time and
cross-sectional dimensions, i.e. addressing potential overlaps in the effects of experience and
mental models, remains a challenge, not least because experience constitutes a key ingredient of
Selecting Legal Advisors in M&As 40
the emergence and development of knowledge structures (Walsh, 1995). By using different
indicators across several organizational domains and levels, we sought to alleviate such
concerns. Yet, we acknowledge the difficulty of appropriately operationalizing multiplicity of
mental models, in particular given the aggregate and indirect nature of our data. While we have
done our best to control for experience, our proxies for multiplicity of mental models may still be
intertwined with experience. Therefore, our empirical results should be interpreted as indicative
of our theory at best. Future research could consider alternative proxies and, using smaller
samples, employ other research designs, such as in-depth qualitative studies (e.g., Kaplan, 2008)
and methods such as causal mapping (e.g., Markoczy, 2001) that have been developed for the
elicitation and quantification of knowledge structures. Such an approach would allow for
incorporating the micro-foundations of organizational cognition, i.e. the social dynamics that
influence the development of shared cognition within organizations (e.g., Kaplan, 2008;
Markoczy, 2001)—a dimension that is absent from this study. Follow-up research could then add
valuable insights by offering a multi-level perspective (cf. Eggers & Kaplan, 2013). For
example, inter-organizational differences at the micro-level in the degree and way that managers
come to hold congruent mental models might translate into distinct levels of the proposed
generalized awareness effect, despite the same number of mental models as assessed here, thus
accounting for some of the remaining variation in our data. Finally, future research could
consider, focusing on performance feedback (Greve, 2003), how a lawyer’s performance,
relative to the acquirer’s aspiration level, interacts with or is influenced by multiplicity of mental
models. Our data did not allow us to pursue such questions, but they emanate from this study and
we hope that they will foster scholarly conversation on organizational learning in complex
corporate domains.
Selecting Legal Advisors in M&As 41
REFERENCES
Acharya, V., Davydenko, S. A., & Strebulaev, I. A. 2012. Cash holdings and credit risk. Review
of Financial Studies, 22: 3572-3609.
Adair, W. L., & Brett, J. M. 2005. The negotiation dance: Time, culture, and behavioral
sequences in negotiation. Organization Science, 16: 33-51.
Argote, L. 2012. Organizational learning: Creating, retaining and transferring knowledge. New
York, NY: Springer Science & Business Media.
Argote, L., & Miron-Spektor, E. 2011. Organizational learning: From experience to knowledge.
Organization Science, 22: 1123-1137.
Argyris, C., & Schön, D. A. 1978. Organizational learning. Reading, MA: Addison-Wesley.
Argyris, C., & Schön, D. A. 1996. Organizational learning II. Reading, MA: Addison-Wesley.
Baker, W. E. 1990. Market networks and corporate behavior. American Journal of Sociology, 96:
589-625.
Barkema, H. G., & Schijven, M. 2008. Toward unlocking the full potential of acquisitions: The
role of organizational restructuring. Academy of Management Journal, 51: 696-722.
Baum, J. A., Rowley, T. J., Shipilov, A. V., & Chuang, Y. T. 2005. Dancing with strangers:
Aspiration performance and the search for underwriting syndicate partners. Administrative
Science Quarterly, 50: 536-575.
Bell, S. T., Villado, A. J., Lukasik, M. A., Belau, L., & Briggs, A. L. 2011. Getting specific
about demographic diversity variable and team performance relationships: A meta-analysis.
Journal of Management, 37: 709-743.
Boone, A. L., & Mulherin, J. H. 2007. How are firms sold? Journal of Finance, 62: 847-876.
Selecting Legal Advisors in M&As 42
Boone, C., Wezel, F. C., Witteloostuijn, A. v. 2013. Joining the pack or going solo? A dynamic
theory of new firm positioning. Journal of Business Venturing, 28: 511-527.
Bromiley, P., & Rau, D. 2016. Social, behavioral, and cognitive influences on Upper Echelons
during strategy process: A literature review. Journal of Management, 42: 174-202.
Cameron, A. C., & Trivedi, P. K. 2005. Microeconometrics: Methods and applications.
Cambridge, U.K.: Cambridge University Press.
Carpenter, M. A., & Frederickson, J. W. 2001. Top management teams, global strategic posture,
and the moderating role of uncertainty. Academy of Management Journal, 44: 533-545.
Chakrabarti, A., & Mitchell, W. 2016. The role of geographic distance in completing related
acquisitions: Evidence from U.S. chemical manufacturers. Strategic Management Journal, 37:
673-694.
Chung, S. A., Singh, H., & Lee, K. 2000. Complementarity, status similarity and social capital as
drivers of alliance formation. Strategic Management Journal, 21: 1-22.
Coates, J. C. 2001. Explaining variation in takeover defenses: Blame the lawyers. California Law
Review, 89: 1301-1422.
Cohen, W. M., & Levinthal, D. A. 1990. Absorptive capacity: A new perspective on learning and
innovation. Administrative Science Quarterly, 35: 128-152.
Coleman, J. S. 1990. Foundations of social theory. Cambridge, MA: Harvard University Press.
Corwin, S. A., & Schultz, P. 2005. The role of IPO underwriting syndicates: Pricing, information
production, and underwriter competition. Journal of Finance, 60: 443-486.
Cyert, R. M., & March, J. G. 1963. A behavioral theory of the firm. Englewood Cliffs, NJ:
Prentice-Hall.
Selecting Legal Advisors in M&As 43
Dahlin, K. B., Weingart, L. R., & Hinds, P. J. 2005. Team diversity and information use.
Academy of Management Journal, 48: 1107-1123.
DePamphilis, D. M. 2008. Mergers, acquisitions, and other restructuring activities (4th ed.). Los
Angeles, CA: Elsevier Academic Press.
Dikova, D., Rao Sahib, P., & Witteloostuijn, A. v. 2010. Cross-border acquisition abandonment
and completion: The effect of institutional differences and organizational learning in the
international business service industry. Journal of International Business Studies, 41: 223-
245.
Duijsters, G., Heimeriks, K. H., Lokshin, B., Meijer, E., & Sabidussi, A. 2012. Do firms learn to
manage alliance portfolio diversity? The diversity performance relationship and the
moderating effects of experience and capability. European Management Review, 9: 139-152.
Eggers, J. P., & Kaplan, S. 2013. Cognition and capabilities: A multi-level perspective. Academy
of Management Annals, 7: 295-340.
Faulconbridge, J., & Muzio, D. 2016. Global professional service firms and the challenge of
institutional complexity: ‘Field relocation’ as a response strategy. Journal of Management
Studies, 53: 89-124.
Fiol, C. M., & Lyles, M. A. 1985. Organizational learning. Academy of Management Review, 10:
803-813.
Flavell, J. 1987. Speculations about the nature and development of metacognition. In F. E.
Weinert & R. H. Kluwe (Eds.), Metacognition, motivation, and understanding: 21-29.
Hillside, NJ: Erlbaum.
Gargiulo, M., & Benassi, M. 2000. Trapped in your own net? Network cohesion, structural holes,
and the adaptation of social capital. Organization Science, 11: 183-196.
Selecting Legal Advisors in M&As 44
Gilson, R. J. 1984. Value creation by business lawyers: Legal skills and asset pricing. Yale Law
Journal, 94: 239-314.
Gilson, R. J., & Black, B. 1996. The law and finance of corporate acquisitions. New York, NY:
Foundation Press.
Ginsberg, A. 1989. Construing the business portfolio: A cognitive model of diversification.
Journal of Management Studies, 26: 417-438.
Glendening, M., Khurana, I. K., & Wang, W. 2016. The market for corporate control and
dividend policies: Cross-country evidence from M&A laws. Journal of International Business
Studies, 47:1106-1134.
Granovetter, M. 1985. Economic action and social structure: The problem of embeddedness.
American Journal of Sociology, 91: 481-510.
Greve, H. R. 2003. Organizational learning from performance feedback: A behavioral
perspective on innovation and change. Cambridge, U.K.: Cambridge University Press.
Gulati, R. 1995. Does familiarity breed trust? The implications of repeated ties on contractual
choice in alliances. Academy of Management Journal, 38: 85-112.
Gulati, R., Gargiulo, M. 1999. Where do interorganizational networks come from? American
Journal of Sociology, 104: 1439-1493.
Haleblian, J., & Finkelstein, S. 1999. The influence of organizational acquisition experience on
acquisition performance: A behavioral learning perspective. Administrative Science
Quarterly, 44: 29-56.
Hambrick, D. C., Davidson, S. C., Snell, S. A., & Snow, C. C. 1998. When groups consist of
multiple nationalities. Organization Studies, 19: 181-205.
Selecting Legal Advisors in M&As 45
Haspeslagh, P. C., & Jemison, D. B. 1991. Managing acquisitions: Creating value through
corporate renewal. New York, NY: Free Press.
Haunschild, P. R., & Sullivan, B. N. 2002. Learning from complexity: Effects of prior accidents
and incidents on airlines’ learning. Administrative Science Quarterly, 47: 609-643.
Hayward, M. L. 2003. Professional influence: The effects of investment banks on clients'
acquisition financing and performance. Strategic Management Journal, 24: 783-801.
Heckman, J. J. 1979. Sample selection bias as a specification error. Econometrica, 47: 153-161.
Heinz, J. P., Nelson, R. L., & Laumann, E. O. 2001. The scale of justice: Observations on the
transformation of urban law practice. Annual Review of Sociology, 27: 337-362.
Hitt, M. A., Dacin, T. M., Levitas, E., Arregle, J., & Borza, A. 2000. Partner selection in
emerging and developed market contexts: Resource-based and organizational learning
perspectives. Academy of Management Journal, 43: 449-467.
Howard, M. D., Withers, M. C., Carnes, C. M., & Hillman, A. J. 2016. Friends or strangers? It
all depends on context: A replication and extension of Beckman, Haunschild, and Phillips
(2004). Strategic Management Journal, 37: 2222-2234.
Hunter, W. C., & Jagtiani, J. 2003. An analysis of advisor choice, fees, and effort in mergers and
acquisitions. Review of Financial Economics, 12: 65-81.
Hylton, K. N., & Deng, F. 2007. Antitrust around the world: An empirical analysis of the scope
of competition laws and their effects. Antitrust Law Journal, 74.2: 271-341.
Jemison, DB, Sitkin, SB. 1986. Corporate acquisitions: A process perspective. Academy of
Management Review, 11(1): 145–163.
Jensen, M. 2003. The role of network resources in market entry: Commercial banks’ entry into
investment banking. Administrative Science Quarterly, 48: 466-497.
Selecting Legal Advisors in M&As 46
Kaplan, S. 2008. Framing contests: Strategy making under uncertainty. Organization Science,
19: 729-752.
Karsten, C. J., Malmendier, U., & Sautner, Z. 2014. M&A negotiations and lawyer expertise.
Working Paper. Available at SSRN 2576866.
Krishnan, C. N. V., Masulis, R. W. 2013. Law firm expertise and merger and acquisition
outcomes. Journal of Law and Economics, 56: 189-226.
La Porta, R., Lopez-de-Silanes, F., & Shleifer, A. 2008. The economic consequences of legal
origins. Journal of Economic Literature, 46: 285-332.
Lang, L. H. P, & Stulz, R. M. 1994. Tobin’s Q, corporate diversification, and firm performance.
Journal of Political Economy, 102: 1248-1280.
Lazega, E. 2001. The collegial phenomenon: The social mechanisms of cooperation among peers
in a corporate law partnership. Oxford, U.K.: Oxford University Press.
Lee, J. J. 2013. Dancing with the enemy? Relational hazards and the contingent value of repeat
exchanges in M&A markets. Organization Science, 24: 1237-1256.
Lim, B. C., & Klein, K. J. 2006. Team mental models and team performance: A field study of the
effects of team mental model similarity and accuracy. Journal of Organizational Behavior,
27: 403-418.
Mackey, T. B., Barney, J. B., & Dotson, J. P. 2017. Corporate diversification and the value of
individual firms: A Bayesian approach. Strategic Management Journal, 38: 322-341.
Marano, V., Arregle, J. L., Hitt, M. A., Spadafora, E., & van Essen, M. van. 2016. Home country
institutions and the internationalization-performance relationship: A meta-analytic review.
Journal of Management, 42: 1075-1110.
Selecting Legal Advisors in M&As 47
March, J. 1991. Exploration and exploitation in organizational learning. Organization Science, 2:
71-87.
Markoczy, L. 1997. Measuring beliefs: Accept no substitutes. Academy of Management Journal,
40: 1228-1242.
Markoczy, L. 2001. Consensus formation during strategic change. Strategic Management
Journal, 22: 1013-1031.
Miller, D. 1996. A preliminary typology of organizational learning: Synthesizing the literature.
Journal of Management, 22: 485-505.
Mitsuhashi, H., & Min, J. 2016. Embedded networks and suboptimal resource matching in
alliance formations. British Journal of Management, 27: 287-303.
Mizruchi, M. S., & Stearns, L. B. 2001. Getting deals done: The use of social networks in bank
decision-making. American Sociology Review, 66: 647-671.
Muehlfeld, K., Rao Sahib, P., & Witteloostuijn, A. v. 2012. A contextual theory of
organizational learning from failures and successes – A study of acquisition completion in the
global newspaper industry, 1981-2008. Strategic Management Journal, 33: 938-964.
Nadolska, A., & Barkema, H. G. 2014. Good learners: How top management teams affect the
success and frequency of acquisitions. Strategic Management Journal, 35: 1483-1507.
Nielsen, B. B., & Nielsen, S. 2013. Top management team nationality diversity and firm
performance: A multilevel study. Strategic Management Journal, 34: 373-382.
Nystrom, P. C., & Starbuck, W. H. 1984. To avoid organizational crises, unlearn. Organizational
Dynamics, 12: 53-65.
Officer, M. S. 2003. Termination fees in mergers and acquisitions. Journal of Financial
Economics, 69: 431–467.
Selecting Legal Advisors in M&As 48
Pennings, J. M., Barkema, H., Douma, S. 1994. Organizational learning and diversification.
Academy of Management Journal, 37: 608-640.
Powell, K. S., & Rhee, M. 2016. Experience in different institutional environments and foreign
subsidiary ownership structure. Journal of Management, 42: 1434-1461.
Powell, W. W., & Smith-Doerr, L. 1994. Networks and economic life. In S. Neil & S. Richard
(Eds.), Handbook of economic sociology: 368-402. Princeton, NJ: Princeton University Press.
Prahalad, C. K., & Bettis, R. A. 1986. The dominant logic: A new linkage between diversity and
performance. Strategic Management Journal, 7: 485-501.
Rouse, W. B., & Morris, N. M. 1986. On looking into the black box: Prospects and limits in the
search for mental models. Psychological Bulletin, 100: 349-363.
Rossi, S., & Volpin, P. F. 2004. Cross-country determinants of mergers and acquisitions. Journal
of Financial Economics, 74: 277-304.
Samuelson, W., & Zeckhauser, R. 1988. Status quo bias in decision-making. Journal of Risk and
Uncertainty, 1: 7-59.
Schraw, G., & Dennison, R. S. 1994. Assessing metacognitive awareness. Contemporary
Educational Psychology, 19: 460-475.
Senge, P. 1990. The fifth discipline: The art and science of the learning organization. NewYork,
NY: Currency Doubleday.
Serdar Dinc, I., & Erel, I. 2013. Economic nationalism in mergers and acquisitions. Journal of
Finance, 68: 2471-2514.
Shah, R. H., Swaminathan, V. 2008. Factors influencing partner selection in strategic alliances:
The moderating role of alliance context. Strategic Management Journal, 29: 471-494.
Selecting Legal Advisors in M&As 49
Simons, D. J., & Chabris, C. F. 1999. Gorillas in our midst: Sustained inattentional blindness for
dynamic events. Perception, 28: 1059-1074.
Spender, J. C. 1987. Industry recipes: An enquiry in the nature and sources of managerial
judgement. Oxford, U.K.: Blackwell.
Stinchcombe, A. L. 1990. Information and organization. Los Angeles, CA: University of
California Press.
Szulanski, G. 1996. Exploring internal stickiness: Impediments to the transfer of best practice
within the firm. Strategic Management Journal, 17: 27-43.
Tosey, P., Visser, M., Saunders, M. N. 2012. The origins and conceptualizations of ‘triple loop’
learning: A critical review. Management Learning, 43: 291-307.
Tate, G., Yang, L. 2015. The bright side of corporate diversification: Evidence from internal
labor markets. Review of Financial Studies, 28: 2203-2249.
Uzzi, B., & Lancaster, R. 2004. Embeddedness and price formation in the corporate law market.
American Sociological Review, 69: 319-344.
Very, P., & Schweiger, D. 2001. The acquisition process as learning process: Evidence from a
study of critical problems and solutions in domestic and cross-border deals. Journal of World
Business, 36: 11-31.
Visser, M. 2007. Deutero-learning in organizations: A review and a reformulation. Academy of
Management Review, 32: 659-667.
Walsh, J. P. 1995. Managerial and organizational cognition. Organization Science, 6: 280-321.
Wan, W. P., Hoskisson, R. E., Short, J. C., & Yiu, D. W. 2011. Resource-based theory and
corporate diversification: Accomplishments and opportunities. Journal of Management, 37:
1335-1368.
Selecting Legal Advisors in M&As 50
Weitzel, U., & Berns, S. 2006. Cross-border takeovers, corruption, and related aspects of
governance. Journal of International Business Studies, 37: 786-806.
White & Case. 2009. Worldwide antitrust merger notification requirements. Washington, DC:
White & Case.
Wiersema, M. F., & Bowen, H. P. 2009. The use of limited dependent variable techniques in
strategy research: Issues and methods. Strategic Management Journal, 30: 679-692.
Wong, P., & O’Sullivan, N. 2001. The determinants and consequences of abandoned takeovers.
Journal of Economic Surveys, 15: 145-186.
Yang, H., Lin, Z., & Peng, M. 2011. Behind acquisitions of alliance partners: Exploratory
learning and network embeddedness. Academy of Management Journal, 54: 1069-1080.
Zhou, C., Xie, J., & Wang, Q. 2016. Failure to complete cross-border M&As: “To” vs. “from”
emerging markets. Journal of International Business Studies, 47: 1077-1105.
Selecting Legal Advisors in M&As 51
FOOTNOTES
1 First, the public announcement presupposes that fundamental issues such as strategic fit have
mostly been resolved (Haspeslagh & Jemison, 1991). Second, announcing a deal entails the
revelation of private information about a bidder’s strategy (Officer, 2003), and disruption of
organizational routines (Jemison & Sitkin, 1986), and puts reputations on the line (Luo, 2005).
Third, Karsten et al. (2014) emphasize agency concerns: Acquirers may prefer a fast closing as
sellers retain control until official consummation, allowing them to extract private benefits.
2 Various factors may trigger a switch between levels of learning: first, factors external to the
learning process such as crises (e.g., Fiol & Lyles, 1985); second, factors inherent to the learning
process such as experience (e.g., Haleblian & Finkelstein, 1999). Third, performance feedback
theory emphasizes their joint influence by focusing on how performance is interpreted relative to
own historical and to social aspiration levels (e.g., Greve, 2003), predicting the existence of a
performance threshold beyond which organizations engage in distant rather than local search.
3 Thus, we assume a substantial degree of collectivity of these group mental models, rather than
investigating the dynamic and interactive social processes that influence the development of
shared cognition within organizations (e.g., Kaplan, 2008, 2011; Markoczy, 1997, 2001).
4 To improve data quality and validity of results, we manually corrected for typos in lawyer
names and country affiliations (by checking the websites), accounted for mergers in the industry,
and removed undisclosed acquirers and acquirers that represent nations or investor groups.
5 As alternatives to focusing on the largest advisor, we (1) analyzed only single-advised deals and
(2) reproduced the analysis at the advisor team level, by calculating the law firm-related variables
based on averages and maxima of the counsels’ attributes, respectively. Results (available upon
request) remained qualitatively unchanged.
Selecting Legal Advisors in M&As 52
Table 1
Summary Statistics of Sample B Pearson pairwise correlations
Variables Mean Std. Dev Min Max 1. 2. 3. 4. 5. 6. 7.
1. Completion time (log(days)) 3.0 1.9 0 8.1
2. Advisor-client ties (#) 1.6 3.5 0 48.0 .01
3. Advisor-client tie str. [0,1] .19 .29 0 1 -.09* .39*
4. Law firm acq. ctr. exp. [0,1] .46 .39 0 1 -.06* .17* .23*
5. Law firm target ctr. exp. [0,1] .33 .37 0 1 -.07* .07* .06* .42*
6. Law firm worldwide exp. (log(#)) 5.3 1.4 0 7.0 .12* .13* .08* -.14* -25*
7. Number of advisors (#) 1.7 1.3 1 12.0 .24* -.03 -.13* -.13* -.19* .29*
8. Acq. internat. M&A exp. (log(#)) 1.2 1.1 0 4.3 .08* .24* -.20* -.19* -.10* .09* .05*
9. (Acq. internat. M&A exp. (log)) 2 2.8 3.5 0 18.9 .07* .25* -.18* -.14* -.09* .09* .04*
10. Acq. network size (log(#)) 1.6 .92 0 4.0 .08* .28* -.19* -.03* -.04* .15* .07*
11. Acq. home ctr. exp. [0,1] .77 .32 0 1 -.04* .02 .19* .19* .13* -.03 -.01
12. Acq. target ctr. exp. [0,1] .14 .24 0 1 .01 .13* .18* .45* .50* -.06* -.06*
13. Acq. mmm antitrust systems [1,10] 2.6 1.6 1 9 .04* .22* -.15* -.14* -.09* .07* .02
14. Acq. mmm political systems [1,10] 3.3 2.0 1 10 .07* .21* -.17* -.15* -.11* .12* .05*
15. Acq. mmm product markets (log(#)) 2.1 .93 0 4.8 .08* .22* -.15* -.04* -.02 .06* .03*
16. Acq. nationalities TMT [0,1] .35 .25 .07 1 -.04* -.01 -.13 -.10* .15 .03 .02
17. Acq. size (log($US)) 9.2 2.4 -3.2 14.9 .18* .12* -.22* -.12* -.04* .08* .01*
18. Public acquirer (0,1) .74 .44 0 1 .04* -.05* .14* -.01 .01 -.01 .03*
19. Deal value (log($US)) 5.2 1.7 2.3 12.2 .45* .05* -.12* -.09* -.07* .25* .43*
20. Number of bidders (#) 1.04 .21 1 4 .09* -.02 -.04* -.01 -.01 .05* .12*
21. Public target (0,1) .30 .47 0 1 .37* -.01 -.08* -.02 .02 .08* .22*
22. Stock payment [0,1] .12 .31 0 1 .17* .03* .08* .13* .06* -.02 .11*
23. Tender offer (0,1) .14 .35 0 1 .21* -.03 -.06* -.07* -.02 .05* .15*
24. Common law ctr. (0,1) .70 .46 0 1 -.01 08* .08* .29* .54* -.01 -.02
25. Same legal origin (0,1) .98 .10 0 1 -.01 .06* .05* .20* .23* -.06* -.05*
8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.
9. (Acq. internat. M&A exp.) 2 .94*
10. Acq. network size .72* .71*
11. Acq. home ctr. exp. -.36* -.34* -.29*
12. Acq. target ctr. exp. -.18* -.17* -.08* .36*
13. Acq. mmm antitrust systems .81* .82* .64* -.34* -.17*
14. Acq. mmm political systems .79* .79* .70* -.33* -.16* .78*
15. Acq. mmm product markets .55* .63* .56* -.25* -.08* .53* .55*
16. Acq. mmm nationalities TMT -.08 -.03 -.12* -.35* -.06 -.06 -.01 .01
17. Acq. Size .58* .58* .51* -.32* -.12* .52* .52* .46* -.02
18. Public acquirer -.29* -.29* -.29* .44* .11* -.27* -.29* -.30* .24 -.26*
19. Deal value .18* .16* .17* -.05 .02 .13* .15* .13* .04 .31* .04*
20. Number of bidders .01 .01 .01 -.02 -.01 -.01 .01 .01 -.01 -.01 -.01*
21. Public target .06* .05* .03* -.05* .07* .03 .03 .06* .21 .01 -.01*
22. Stock payment -.11* -.09* -.13* .14* .19* -.09* -.14* -.11* -.10 -.06* .20*
23. Tender offer .05* .04* -.01 -.09* -.03* .03 .01 .04* .30 -.01 .04*
24. Common law ctr. -.12* -.11* -.04* .10* .40* -.13* -.10* .01 .19 -.08* .01
25. Same legal origin -.10* -.09* -.03* .17* .42* -.09* -.08* -.03* -.08 .04* .05*
19. 20. 21. 22. 23. 24.
20. Number of bidders .12*
21. Public target .18* .21*
22. Stock payment .18* .03 .24*
23. Tender offer .08* .17* .59* -.01
24. Common law ctr. -.01 .02 .09* .08* .01
25. Same legal origin .01 .01 .06* .11* .02 .31*
No. announced mergers = 11,511
NOTES: (a) * Significance level: p<.001. (b) Pairwise correlations for acquirer nationalities TMT only available for a sample of
65 law firm-merger dyads in 2010.
Selecting Legal Advisors in M&As 53
Table 2
Results of the Selection Models Selection equations: Probability of selecting a law firm
Expected
sign
Domestic
mergers
(1)
Internat.
mergers
(2)
Internat.
mergers:
experiential
learning
(3)
Internat.
mergers:
antitrust
systems
exposure
(4)
Internat.
mergers:
politcial
systems
exposure
(5)
Internat.
mergers:
Product
markets
exposure
(6)
Mental model second-order interactions
Mental model variable X
Experiential learning int.3 (MM1) (H4: -) -.02 (.072) -.01 (.317) -.10 (.000)
Experiential learning int.4 (MM2) (H4: +) .01 (.002) .01 (.000) -.01 (.529)
Experiential learning interactions
Acq. cross-border M&A exp. (log) X
Advisor-client tie strength (EL1) (H3: +) .31 (.002) .21 (.059) .20 (.059) .10 (.388)
Law firm target ctr. exp. (EL2) (H3: -) -.05 (.165) .02 (.685) .03 (.372) -.04 (.326)
(Acq. cross-border M&A exp. (log))2 X
Advisor-client tie strength (EL3)
(H3: -)
-.07 (.121)
.04 (.565)
.14 (.821)
.26 (.000)
Law firm target country exp. (EL4) (H3: +) .03 (.025) -.03(.170) -.03 (.052) .01 (.452)
Exclusion restriction
Law firm worldwide experience (log) X
Avg. law firm exp.in same country-year
(+)
.01 (.000)
.01 (.000)
.01 (.000)
.01 (.000)
.01 (.000)
.01 (.000)
Law firm variables
Advisor-client ties (H1: +) .06 (.000) .09 (.000) .09 (.000) .09 (.000) .09 (.000) .10 (.000)
Advisor-client tie strength (H1: +) 2.08 (.000) 2.02 (.000) 1.90 (.000) 1.90 (.000) 1.90 (.000) 1.88 (.000)
Acquirer country expertise (H1: +) .77 (.000) .20 (.000) .20 (.000) .20 (.000) .21 (.000) .20 (.000)
Target country expertise (H1: +) --- .66 (.000) .66 (.000) .65 (.000) .64 (.000) .65 (.000)
Worldwide experience (log) (+) 32 (.000) .21 (.000) .20 (.000) .21 (.000) .20 (.000) .21 (.000)
Number of advisors (+) .20 (.000) .17 (.000) .17 (.000) .17 (.000) .17 (.000) .17 (.000)
Acquirer variables
Cross-border M&A exp. (log) (-) .09 (.000) .02 (.031) .02 (.200) .02 (.193) .02 (.173) .02 (.347)
(Cross-border M&A exp. (log))2 (-) -.03 (.000) -.02 (.000) .02 (.001) -.01 (.004) -.01 (.001) -.02 (.002)
Network size (log) (?) -.10 (.000) -.02 (.006) -.02 (.017) -.02 (.015) -.02 (.009) -.02 (.014)
Home country experience (-) -.17 (.000) -.04 (.023) -.04 (.024) -.04 (.022) -.04 (.034) -.05 (.012)
Target country experience (-) --- -.16 (.000) -.16 (.000) -.16 (.000) -.15 (.000) -.16 (.000)
Acq. mmm antitrust systems (-) --- -.01 (.398) -.01 (.455) -.01 (.149) --- ---
Acq. mmm political systems (-) --- --- --- --- -.01 (.866) ---
Acq. mmm product markets (log) (-) --- --- --- --- --- .02 (.226)
Acquirer size (log) (?) -.01 (.475) .01 (.357) .01 (.376) .01 (.325) -.01 (.307) -.00 (.689)
Public acquirer (?) .01 (.319) -.01 (.609) -.01 (.555) -.01 (.587) -.01 (.780) -.01 (.348)
Announced mergers (sample B) 5,594 5,917 5,917 5,917 5,680 5,917
Law firm-mergers dyads (sample C) 524,506 634,669 634,669 634,669 613,482 634,669
NOTES: (a) The table reports point estimates and p-values in parentheses. (b) All models include a constant, 7 merger control
variables introduced in the methodology section, and 12 year dummies (not reported). (c) Models 4-6 omit two potential
interaction terms between the multiplicity of mental model variables and the linear experience interactions, EL1 and EL2. The
omission has solely a statistical reason, since these “linear” interaction terms are highly correlated with MM1 and MM2
(P.C.>.95). Nevertheless, from a logical viewpoint, either set of interaction terms is sufficient to test H4, because both shift the
turning point of the experiential learning parabola. The question still is whether our findings are robust to that choice. In online
appendix 8 Table A8, we successfully reproduce the results presented above using the alternative set of mental model interaction
terms.
Selecting Legal Advisors in M&As 54
Table 3
Results of the Outcome Models
Outcome equations: Duration of closing phase (log)
Expected
sign
Heckman:
Domestic
mergers
(7)
OLS:
International
mergers
(8)
OLS:
Domestic
mergers
(9)
OLS:
International
mergers
(10)
Law firm variables
Advisor-client ties (H2: -) .01 (.065) .01 (.530)
Advisor-client tie strength (H2: -) -.36 (.000) .01 (.889)
Acquirer country expertise (H2: -) -.05 (.512) -.21 (.018)
Target country expertise (H2: -) --- -.33 (.000)
Worldwide experience (log) (-) .02 (.311) .04 (.038) -.02 (.323) .02 (.369)
Number of advisors (+) -.01 (.885) .03 (.138) .01 (.998) .02 (.271)
Acquirer variables
Cross-border M&A exp. (log) (-) -.05 (.371) -.02 (.734) -.07 (.240) -.03 (.603)
(Cross-border M&A exp. (log))2 (-) .01 (.699) -.02 (.399) .01 (.594) -.02 (.380)
Network size (log) (-) .02 (.579) -.01 (.975) .01 (.843) .02 (.698)
Home country experience (-) -.03 (.767) -.14 (.148) -.03 (.718) -.10 (.282)
Target country experience (-) --- .02 (.826) --- .05 (.660)
Acq. mmm antitrust systems (?) --- -.03 (.209) --- -.03 (.194)
Acquirer size (log) (+) .03 (.022) .07 (.000) .03 (.036) .07 (.000)
Public acquirer (+) .14 (.013) .13 (.040) .15 (.011) .13 (.037)
Merger variables
Deal value (log) (+) .39 (.000) .41 (.000) .39 (.000) .41 (.000)
Number of bidders (+) .05 (.708) .01 (.934) .06 (.614) .02 (.909)
Public target (+) 1.08 (.000) .81 (.000) 1.08 (.000) .80 (.000)
Stock payment (+) .39 (.000) .38 (.000) .39 (.000) .38 (.000)
Tender offer (+) .02 (.814) .20 (.035) .02 (.817) .20 (.033)
Common law target country (-) -.10 (.138) -.11 (.045) -.10 (.156) -.01 (.977)
Same legal origin (-) --- -.030 (.544) --- -.05 (.276)
Announced mergers (Sample B) 5,594 5,917 5,594 5,917
Law firm-mergers dyads (Sample C) 524,506 634,669 524,506 634,669
Inverse Mills Ratio .13 (.001) .04 (.383) .49 (.124) -.33 (.536)
NOTES: (a) The table reports point estimates and p-values in parentheses. (b) The p-values of the Inverse Mills
Ratio stem from a Wald test for independence of the outcome equations (Models 7-10) from the corresponding
selection equations (Models 1 and 2). H0: Coefficient of the Inverse Mills Ratio=0. The H0 is not rejected for
Models 8-10. Thus, as Heckman selection models produce inefficiently high standard errors, we present, as
recommended by Cameron & Trivedi (2005: 552), the coefficients and p-values of a standard OLS regression. (c)
All models include a constant and year dummies (not reported). (d) Models 8 and 10 show that an acquirer’s
multiplicity of mental models, as measured by the variable mmm antitrust systems, has in itself no direct effect on
duration. This also holds for the other two mental model proxies mmm political systems and mmm product markets
(unreported regressions, available upon request).
Selecting Legal Advisors in M&As 55
Figure 1
The Moderating Effect of M&A Experience and Multiplicity of Mental Models
NOTES: (a) The panels present: ME(advisor-client tie strength) - ME(law firm target country expertise) on the likelihood of
selection, at different levels of acquirer cross-border M&A experience and multiplicity of mental models (mmm). ME stem from
Models 4-6 (Table 2) and are illustrated by quadratic prediction plots. (b) The values for the mmm variables are chosen so as to
highlight the emergence and inwards shift of the turning point. The bottom panel, e.g., shows that no turning point is in reach for
acquirers in the bottom 50% of mmm product markets. A turning point emerges for acquirers in the 6th decile and is shifted to the
left in the 8th decile.
.03
.04
.05
.06
.07
Marg
ina
l E
ffect(
tie s
treng
th)-
ME
(ta
rget ctr
. exp)
0 1 2 3 4Bidder cross-border M&A experience (log)
MMM Antitrust 3rd decile MMM Antitrust 4th decile
MMM Antitrust 6th decile
.04
.045
.05
.055
.06
.065
Marg
ina
l E
ffect(
tie s
treng
th)-
ME
(ta
rget ctr
. exp)
0 1 2 3 4Bidder cross-border M&A experience (log)
MMM Stability 6th decile MMM Stability 7th decile
MMM Stability 8th decile
.03
.04
.05
.06
.07
.08
Marg
ina
l E
ffect(
tie s
treng
th)-
ME
(ta
rget ctr
. exp)
0 1 2 3 4Bidder cross-border M&A experience (log)
MMM Products 5th decile MMM Products 6th decile
MMM Products 8th decile